Compact Letter Display (CLD) for Tukey HSD Results

📃 Overview

In statistical analysis, particularly in agricultural, biological, and social sciences research, comparing group means is essential for understanding treatment effects. Tukey’s Honest Significant Difference (HSD) test is widely used for post-hoc comparisons after ANOVA. However, interpreting results can be challenging when multiple groups are involved.

This project implements a Compact Letter Display (CLD) system in Python to simplify Tukey HSD results by assigning letters to groups. Groups that share the same letter are not significantly different, while groups with different letters are significantly distinct. This approach provides a clear, concise, and publication-ready interpretation of Tukey’s HSD output.

---

⚙️ Workflow / Methodology

1. Data Preparation

   • Load dataset into a Pandas DataFrame.
   • Identify the dependent variable and grouping factor.

2. Perform ANOVA

   • Conduct one-way ANOVA to determine if significant differences exist.

3. Run Tukey’s HSD Test

   • Apply statsmodels Tukey HSD procedure to compute pairwise group comparisons.

4. Generate Compact Letter Display (CLD)

   • Process Tukey HSD results.
   • Assign letters to each group based on significance.

5. Output Results

   • Generate a clean summary table with means, standard errors, and assigned letters.
   • Provide easy-to-interpret visualizations (optional).

---

🛠️ Technology Used

• Python (Core programming language)
• Pandas → Data manipulation and handling
• NumPy → Numerical computations
• SciPy → ANOVA and statistical testing
• Statsmodels → Tukey HSD implementation
• Matplotlib / Seaborn → Visualization of results (barplots with letters)

---

📈 Results

• Groups are displayed alongside their means, standard errors, and CLD letters.
• Compact letters make it easy to interpret which groups differ without scanning through large pairwise comparison tables.
• Visual representation provides a clearer understanding of treatment effects for reports, academic papers, or presentations.

Compact Letter Display (CLD) Table:

(Groups sharing the same letter are not significantly different.)

---

🌍 Applications

• Agricultural Research → Comparing crop yields under different treatments.
• Biological Experiments → Analyzing gene expression, drug effects, or ecological differences.
• Social Sciences → Simplifying survey or behavioral study results.
• Publication & Reporting → Generating publication-ready outputs with minimal manual editing.

---

🚀 Usage

1. Clone this repository:

   git https://github.com/Jabulente/Compact-Letter-Display.git

2. Install dependencies:

   pip install -r requirements.txt

3. Run the example script:

   python main.py

4. Replace the dataset with your own experimental data.

---

🤝 Contribution

Contributions are welcome! If you’d like to improve the workflow, add new visualization options, or extend the project to other post-hoc tests, feel free to fork the repo and submit a pull request.

---

📌 License

This project is licensed under the MIT License.